The Five Eyes intelligence alliance, comprising the US, UK, Canada, Australia, and New Zealand, issued a joint warning on June 22, 2026, about the rapid increase in cyberattack threats powered by advanced AI models. The alliance said cyber threat evolution will take place within months instead of years due to AI lowering technical barriers and increasing attack speed and complexity. [1, 2, 3]

The statement stressed that breaches will become inevitable without preparedness. "The rapid pace of frontier AI development means cyber risk assumptions can become outdated in months, not years. Breaches will occur. Preparedness helps you contain them quickly and prevent escalation into major operational and financial crises," the Five Eyes joint statement said. [1]

To meet this challenge, the alliance urged governments and enterprises to strengthen cybersecurity defenses urgently. Recommended actions include integrating AI tools for vulnerability detection, patching software promptly, and restricting unnecessary system access and network connections. [1, 2, 3]

Anthropic, a US AI firm, has developed advanced models Mythos and Fable 5, which can quickly identify software vulnerabilities and assist in sophisticated cyberattacks. These models succeeded in bypassing Apple operating systems in 6 out of 10 tests, raising national security concerns. [1, 2, 3, 4]

Due to these concerns, the US government ordered a restriction on foreign nationals' access to Mythos 5 and Fable 5 in early June 2026, forcing Anthropic to suspend external access to these AI models. [1, 2, 3, 5]

In response to rising AI-driven threats, the US Cybersecurity and Infrastructure Security Agency (CISA) shortened government vulnerability patch deadlines to three days to improve defense readiness. [2, 3, 6]

While the Five Eyes statement did not name specific countries, some analysts suspect China and Russia are actively developing AI-enhanced cyberattack capabilities. China's AI company Beijing Zhipu Technology released GLM-5.2 on June 17, 2026, a top global large language model. Its stock price increased approximately 20 times in six months, reaching HKD 2,980 on June 22, with a market valuation of HKD 1.1 trillion, signaling a narrowing technology gap with US models. [7, 8, 9, 5]

US experts warned against complacency. Alex Stamos, former Facebook Chief Security Officer, said, "China may have quietly developed highly competitive AI technology, and US complacency from longstanding leadership is a dangerous arrogance." Olivia Shen from the University of Sydney noted, "The next powerful AI model like Mythos or Fable may emerge soon. We cannot focus only on currently public models because stronger ones are likely being developed worldwide." [8, 5]

The US internal debate on AI risk standards has slowed unified cybersecurity measures, but CISA continues stepping up enforcement despite challenges. [8, 4]

Also on June 22, US President Trump signed executive orders to accelerate quantum computing and strengthen defenses against quantum-enabled cyber threats. "We have already been far ahead [in technology], now we will further expand that lead," he said. [7]

The Five Eyes warning highlights an urgent need for accelerated cybersecurity efforts, including AI-powered defenses and rapid patching. The alliance's statement and related government actions mark a critical response within weeks of recognizing the AI-driven cyber risk surge.